SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.02353 · AAAI · 2023

Self-Asymmetric Invertible Network for Compression-Aware Image Rescaling

Li Zhang, Shijie Zhao, Jinhai Yang, Junlin Li, Mengxi Guo

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (4 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2303.02353.json · badge: /api/ran/2303.02353.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
yang-jin-hai/SAIN community
Papers with Code archive
4 ran · 1 no recorded run
FunctionStatusWhere it lives
EnhancedInvBlock Ran
ran on a synthesized input · constructed an object; no result was computed
yang-jin-hai/SAIN/codes/models/modules/inv_arch.py
code served (permissive licence) · get_code("c48aa122343168e6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, EnhancedInvBlock: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Define a simple subnet constructor that takes input and output channels
class SimpleSubnet(nn.Module):
    def __init__(self, in_channels, out_channels):
        super(SimpleSubnet, self).__init__()
        self.net = nn.Sequential(
            nn.Linear(in_channels, 8),
            nn.ReLU(),
            nn.Linear(8, out_channels)
        )
    
    def forward(self, x):
        # x shape: (batch, channels, height, width) or (batch, channels)
        # Flatten spatial dimensions if they exist
        if x.dim() == 4:
            batch, channels, h, w = x.shape
            x = x.view(batch, channels, -1).permute(0, 2, 1)  # (batch, h*w, channels)
            x = self.net(x)
            x = x.permute(0, 2, 1).view(batch, -1, h, w)
            return x
        else:
            return self.net(x)

def subnet_constructor(in_channels, out_channels):
    return SimpleSubnet(in_channels, out_channels)

# Parameters for EnhancedInvBlock
channel_num = 8
channel_split_num = 3
clamp = 1.0

ARGS = {
    'subnet_constructor': subnet_constructor,
    'channel_num': channel_num,
    'channel_split_num': channel_split_num,
    'clamp': clamp
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EnhancedInvBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a EnhancedInvBlock output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
GaussianMixture Ran
ran on a synthesized input · constructed an object; no result was computed
yang-jin-hai/SAIN/codes/models/modules/inv_arch.py
code served (permissive licence) · get_code("8bcb0eacf71fa11c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, GaussianMixture: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# GaussianMixture expects gmm_components as an integer parameter
# Looking at the __init__, it creates a parameter tensor of shape (3, gmm_components)
# The refresh() method unpacks this into pis, mus, logvars
# The build() method uses softmax, exp, clamp on these
# The sample() method calls build() and uses the gmm distribution

ARGS = {
    'gmm_components': 5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type GaussianMixture — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a GaussianMixture output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
HaarDownsampling Ran
ran on a synthesized input · constructed an object; no result was computed
yang-jin-hai/SAIN/codes/models/modules/inv_arch.py
code served (permissive licence) · get_code("6638ca9310bdc7bb")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, HaarDownsampling: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import numpy as np

torch.manual_seed(0)

# HaarDownsampling only takes channel_in as a parameter
# channel_in should be a positive integer representing the number of input channels

channel_in = 3

ARGS = {
    'channel_in': channel_in
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type HaarDownsampling — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a HaarDownsampling output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
VanillaInvBlock Ran
ran on a synthesized input · constructed an object; no result was computed
yang-jin-hai/SAIN/codes/models/modules/inv_arch.py
code served (permissive licence) · get_code("edd575de9595285c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, VanillaInvBlock: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Define a simple subnet constructor that takes input_channels and output_channels
class SimpleSubnet(nn.Module):
    def __init__(self, in_channels, out_channels):
        super(SimpleSubnet, self).__init__()
        self.net = nn.Sequential(
            nn.Conv1d(in_channels, 8, kernel_size=1),
            nn.ReLU(),
            nn.Conv1d(8, out_channels, kernel_size=1)
        )
    
    def forward(self, x):
        return self.net(x)

def subnet_constructor(in_channels, out_channels):
    """Factory function to create a subnet"""
    return SimpleSubnet(in_channels, out_channels)

# Create arguments
channel_num = 8
channel_split_num = 3

ARGS = {
    'subnet_constructor': subnet_constructor,
    'channel_num': channel_num,
    'channel_split_num': channel_split_num,
    'clamp': 1.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type VanillaInvBlock — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a VanillaInvBlock output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
SAINet Not yet run
unverified (harvested, no recorded run)
yang-jin-hai/SAIN/codes/models/modules/inv_arch.py
code served (permissive licence) · get_code("4b2376db15a9ca0c")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 later) and 4 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
HaarDownsampling
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2104.10546 (2021-04)
Invertible Denoising Network: A Light Solution for Real Noise Removal

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

High-resolution (HR) images are usually downscaled to lowresolution (LR) ones for better display and afterward upscaled back to the original size to recover details. Recent work in image rescaling formulates downscaling and upscaling as a unified task and learns a bijective mapping between HR and LR via invertible networks. However, in real-world applications (e.g., social media), most images are compressed for transmission. Lossy compression will lead to irreversible information loss on LR images, hence damaging the inverse upscaling procedure and degrading the reconstruction accuracy. In this paper, we propose the Self-Asymmetric Invertible Network (SAIN) for compression-aware image rescaling. To tackle the distribution shift, we first develop an end-toend asymmetric framework with two separate bijective mappings for high-quality and compressed LR images, respectively. Then, based on empirical analysis of this framework, we model the distribution of the lost information (including downscaling and compression) using isotropic Gaussian mixtures and propose the Enhanced Invertible Block to derive high-quality/compressed LR images in one forward pass. Besides, we design a set of losses to regularize the learned LR images and enhance the invertibility. Extensive experiments demonstrate the consistent improvements of SAIN across various image rescaling datasets in terms of both quantitative and qualitative evaluation under standard image compression formats (i.e., JPEG and WebP). Code is available at https://github.com/yang-jin-hai/SAIN.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2303.02353")
get_code_for_paper("2303.02353")
have("2303.02353")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2303.02353.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2303.02353.svg)](https://syntology.ai/paper/2303.02353)

Connect an agent — have() is free.